plot.spheremap {MDSMap}

Produces diagnostic plots for the estimated map using calc.maps.sphere.

Description

Produces diagnostic plots for the estimated map using calc.maps.sphere.

Usage

## S3 method for class 'spheremap'
plot(x, displaytext = TRUE, ...)

Arguments

x

Map object from calc.maps.sphere

displaytext

Logical argument determining how markers should be labelled in the MDS configuration plot. If TRUE then marker names are used. If FALSE then numbers are used.

...

Further arguments are ignored. (accepted for compatibility with generic plot)

Details

Produces a figure with 3 panels from a map object produced by calc.maps.sphere.

Panel one shows the stress of the unconstrained MDS, the stress of the constrained MDS and the ratio of the two. A good rule of thumb is the stress from the constrained MDS should not be more than 10

Panel 2 shows the final configuration of the unconstrained MDS which can be used to identify outliers.

Panel 3 shows the final configuration of the constrained MDS in black and the unconstrained MDS in red. This can be used to check that the constrained fit is not distorting the data - large changes in the rank of a point in either dimension 1 or dimension 2 are indications of a problem with the fit.

Panel 4 shows the pointwise nearest neighbour fits in order of the position in the estimated map.

If D1lim or D2lim is not specified, then limits of panels 2 and 3 are defined by plot.smacof.

Markers are assigned numbers according to the order in which they occur in the input file. The locikey output of the map object is a data frame associating marker names with their numbers. This can be accessed using pcmap3d$locikey. If displaytext=FALSE then in panels 2 and 3 markers will be labelled by these numbers. By default displaytext=TRUE and markers are labelled by marker name.

References

de Leeuw J, Mair P (2009) Multidimensional scaling using majorization: SMACOF in R. J Stat Softw 31: 1-30 http://www.jstatsoft.org/v31/i03/

See Also

plot.pcmap, plot.pcmap3d, plot.smacof, calc.maps.sphere

Examples

map<-calc.maps.sphere(system.file("extdata", "lgI.txt", package="MDSMap"),
weightfn='lod', mapfn='kosambi')
plot(map)